UK Welfare Reform 1996 to 2008 and beyond: A personalised and responsive welfare system? Paul Gregg. April 2008 Working Paper No.

Transcription

1 THE CENTRE FOR MARKET AND PUBLIC ORGANISATION The Centre for Market and Public Organisation (CMPO) is a leading research centre, combining expertise in economics, geography and law. Our objective is to study the intersection between the public and private sectors of the economy, and in particular to understand the right way to organise and deliver public services. The Centre aims to develop research, contribute to the public debate and inform policy-making. CMPO, now an ESRC Research Centre was established in 1998 with two large grants from The Leverhulme Trust. In 2004 we were awarded ESRC Research Centre status, and CMPO now combines core funding from both the ESRC and the Trust. Centre for Market and Public Organisation Bristol Institute of Public Affairs University of Bristol 2 Priory Road Bristol BS8 1TX Tel: (0117) Fax: (0117) cmpo-office@bristol.ac.uk UK Welfare Reform 1996 to 2008 and beyond: A personalised and responsive welfare system? Paul Gregg April 2008 Working Paper No. 08/196 ISSN X

2 CMPO Working Paper Series No. 08/196 UK Welfare Reform 1996 to 2008 and beyond: A personalised and responsive welfare system? Paul Gregg CMPO and Department of Economics, University of Bristol April 2008 Abstract The UK welfare system has undergone three very profound periods of reform of the post-war model laid down by Beveridge. The first was a move in the direction of (but never fully converged with) the Bismarkian model of a contributory social insurance model with time limited earnings related benefits with a low value means tested social assistance safety net. This occurred slowly through the 1960s and up to the mid-1970s. The second phase started in 1979 and involved a dramatic move to curtail the social insurance entitlements and end all earnings related benefits. The result was a residualist low value means tested social assistance model, which ended both the Beveridge model and completely reversed the drift toward a European Bismarkian approach. Finally from 1996 a new model has emerged based on an activational welfare model with greater emphasis on incentives, support services and conditionality. As a direction of travel from the previous regime(s) this represents an increase in the engagement and support functions, increases in the (disciplinary) required activity functions combined with increased financial support for children and pensioners and personalised support services. The emerging model is far from completion and the final make up of the system remains uncertain. However, it bears strong similarities with developments in New Zealand and to a degree Australia and Canada. Within Europe the model most closely resembles a less generous version of the welfare systems in Denmark or Holland, which are sometimes referred to as embodying Flex-security. This evolutionary process of reform had some antecedents prior to 1996 but has really come to the fore since that date. This paper discusses reform in depth from 1996 and looks at its current direction of evolutionary change. Keywords: welfare reform, tax credits, lone parents, disabled adults. JEL Classification: H21, J22, J13, I38 Electronic version: Acknowledgements This report has been prepared for the Friederick Eberts Foundation. Thanks go to Carol Propper and Sarah Smith for the help with the preparation of this paper. Address for Correspondence CMPO, Bristol Institute of Public Affairs University of Bristol 2 Priory Road Bristol BS8 1TX p.gregg@bristol.ac.uk CMPO is jointly funded by the Leverhulme Trust and the ESRC

3 UK Welfare Reform 1996 to 2007 and beyond: A personalised and responsive welfare system? Introduction The UK welfare system has undergone three very profound periods of reform of the post-war model laid down by Beveridge. The first was a move in the direction of the Bismarkian model of a contributory social insurance model with time-limited earnings-related benefits with a low value means tested social assistance safety net. This occurred slowly through the 1960s and up to the mid-1970s. The second phase started in 1979 and involved a dramatic move to curtail the social insurance entitlements and end all earnings-related benefits. The result was a residualist low value means tested social assistance model, which ended both the Beveridge model and completely reversed the drift toward Europe s Bismarkian approach. Finally, from 1996 a new model has emerged based on an activational welfare model with greater emphasis on incentives, support services and conditionality. As a direction of travel from the previous regime(s) this represents an increase in the engagement and support functions, increases in the (disciplinary) required activity functions combined with increased financial support for children and pensioners and personalised support services. The emerging model is far from completion and the final make up of the system remains uncertain. However, it bears strong similarities with developments in New Zealand and to a degree Australia and Canada. Within Europe the model most closely resembles a less generous version of the welfare systems in Denmark or Holland, which are sometimes referred to as embodying Flex-security. This evolutionary process of reform had some antecedents prior to 1996, but has really come to the fore since that date. As a result, this paper discusses reform in depth from 1996 and looks at its current direction of change with a rather more cursory discussion of the demise of contributory social insurance model. 1. The Post-war UK Welfare System

4 1.1 The Beveridge Plan 1942 Although the origins of the British welfare started prior to the Second World War, it was with the Beveridge plan of 1942 that the modern welfare state came into being. Already a century ago all political parties agreed that the state would have to play a bigger role in providing pensions and welfare for the poor. Among the key measures introduced were old age pensions, a system of health insurance, and labour exchanges to help tackle unemployment. However, the Edwardian welfare state was incomplete. It was limited to the working class, its funding basis was insecure, and little progress was made in tackling poverty among people of working age. The end of the WWI also brought a slump, particularly in northern industrial towns, that deepened into the Great Depression by the l930s. By the time Britain entered the Second World War, the pressures for social reform were mounting. In 1942 the Beveridge Report laid out a comprehensive programme of social reform "from the cradle to the grave". His report called for a universal flatrate low value benefit payable to all, on the basis of fixed national insurance contributions. Beveridge wanted to tackle what he called the "five giants" - want, disease, squalor, ignorance, and idleness - through a universal welfare state which would also provide a comprehensive health service, vastly expanded public housing, and free and universal secondary education. Welfare benefits were to cover old age, unemployment, and sickness, as well as benefits for the poor and family allowances. The new Labour government, which took power in 1945, was keen to implement the Beveridge report in full as soon as possible. But, in the severe conditions of post-war Britain, it did not have the money to pay an adequate level of flat-rate benefits that would keep people out of poverty. So from early on, the residual National Assistance (later Supplementary Benefit and now Income Support) played a bigger role in alleviating poverty than Beveridge had planned. Family allowances were also never implemented in the generous way Beveridge had proposed, but they did become more generous in the 1970s as Child Benefit. The idea of basing entitlement on contributions meant that many people, especially women, were excluded from the system. The male bread-winner nuclear family and full employment - and the relative prosperity that went with it - were key to Beveridge's plan.

5 By the mid-1970s the post-war model had evolved from the Beveridge plan into a system closer to a Bismarkian European model with earnings-related payments for unemployment, sickness and pensions, but with time limits on payments for unemployment. However, the gender bias in support through contributions which excluded part-time jobs with shorter hours, as well as those looking after the family, lead to the pressures to widen the net to marginal groups, thus weakening the insurance principle. The long period of Conservative government ( ) saw a steady erosion of the Social Insurance model, with all earnings-related links removed, except for second stage pensions. Yet even here people were given incentive to opt out of the state second stage pension into private pension saving instead. The exemption from means testing for unemployment benefits was also reduced to 6 months. The benefit system thus became a residualised flat-rate system with large reliance on meanstested social assurance benefits (Income Support). The levels of these benefits were fixed in real terms so that their value relative to earnings progressively fell and Child Benefits were often not even uprated with prices. 1.2 The Role of Government, Business and Trade Unions The Beveridge Report, after some resistance from the Conservative party, was implemented by the incoming 1945 Labour government and formed a broad political consensus. The National Insurance system was operated fully by the government as another form of revenue raising and expenditure counterpart. It was never created as a fund separated from government revenues and hence had no overseeing body representing the social partners. In addition to no direct role for employers and trades unions, the system was operated entirely by central government with no role for local or regional government. However, there were tripartite bodies setting minimum wages at the industry level (Wages Councils, see Dickens et al. 1993) and through the 1960s and 70s a number of tripartite bodies grew up discussing economic performance, wage setting, etc. This era was characterised as that of beer and sandwiches, as union leaders were regularly meeting government ministers over threatened strikes and many other issues. Trades unions often campaigned successfully for a number of extensions to working rights, including reductions in working hours, redundancy payments and even limited worker

6 representation. However, in the Conservative era all tripartite bodies were abolished, including the setting of minimum pay levels (in 1993) and restrictions placed on the ability of trade unions to hold strike action and support other workers involved in disputes. Trade union membership and influenced plummeted after Welfare Dependency and Social Expenditure In 1979 around 2 million working age adults were claiming major welfare benefits (unemployment, long-term sickness/disability and lone parenthood). Social expenditures covering these benefits, universal child assistance (Child Benefit) and pensions ran to around 7% of GDP. However, from 1979 the value of the benefits received remained broadly frozen in real terms, and so relative poverty rose steadily among those without work. Yet total expenditures rose as a proportion of GDP to around 13.5% in 1994 (see Figure 1 for breakdown of expenditures). This mainly occurred because of the dramatic rise in welfare dependency associated with the growing number of jobless households (Gregg and Wadsworth, 1999). This rose so dramatically that by 1994 the numbers of working age welfare claimants had reached 6 million, before falling back to 5.5 million by 1997 (see Figure 2). The largest increase was for claims of sickness and disability benefits, with a rise of nearly 2 million. Not only had dependency risen dramatically, but the poverty gap among those on welfare (and in work) had risen dramatically. The proportions of working age adults and children living on incomes below 60% of the median (BHC) more than doubled between 1979 and But rises in the proportion below 50%, 40% and even 30% grew even faster. People were falling much deeper into poverty before the safety net held them.

7 Figure 1 Share of GDP Spent on Welfare Benefits (including Pensions) and Benefit expenditure breakdown Percent of GDP Tax credits Children Working age Pensioners / / / / / / / / / / / / /2004 Year Figure 2 Numbers of Claimants of Major Groupings of Welfare Benefits Targeted on those Not in Work 1995 to 2006 Claimants of benefits Number of claimants / / / / / / / / / / /2006 Year Income support - SDA & DIS Income support - others Income support - lone parents JSA claimants

8 Key aggregate economic figures The UK was characterised as the sick man of Europe in the 1960s and 70s. Average economic growth and productivity was low by European standards, and hence living standards grew more slowly. GDP and productivity Inflation was also relatively high from the late 1960s, peaking in the mid-1970s at 25%. Unemployment was low on current ILO definitions around 4% and employment high, with 60% of the adult population in work or 75% of the working age population aged for men and for women). This period was also one of widespread conflict in industrial relations with a lot of strike action, which was at its most extreme in the Winter of Discontent 1978/79. Inflation 1960 to 2006 (Retail Price Index) Average annual inflation Average GDP Average output per worker Average annual inflation Key years Key years % % % % % % % % % The 1980s saw the economy yawing from recession in , a overheated boom and recession again Over the period as a whole unemployment was very high, as was inflation, although less than in the 1970s. Growth bottomed out at around 2%, which was no longer below that of other European nations, owing to the more rapid slowdown in continental Europe. Employment recovered to that of 1968 and 1975 by 1990, but fell away in the next recession. Employment rose amongst women who were primarily second earners within families, whilst the number of welfare dependent families continued to rise, peaking at 1 in 5 working age households Employment and unemployment 1970 to 2006 Unemployment proportion of economic active Employment as percentage of total population 16+ Unemployment proportion of economic active Employment as percentage of total population 16+ Key years Key years % 61% % 57% % 60% % 60% % 55% % 60% % 60%

9 2. Social Challenges and Welfare Reform from 1996 The economy in the UK has performed well since 1995, being described by the OECD as the Goldilocks economy, neither too hot nor too cold but just right. Growth has been good, averaging 2.8% pa and has had little annual variation (the range has been 1.6% to 4%). This has been the highest among the major G7 economies over this period. Inflation has been low and stable since the Bank of England set interest rates independently of the government. The Retail Price Index measure has risen at around 3%, but the CPI (which is comparable with European measures) has averaged just over 2%. This benign picture led to budget surpluses being generated in , but a move to increase spending on Health and Education very rapidly after the 2001 election has seen the government budget move into sustained and structural deficits. The government is now seeking to reduce these deficits with smaller increases in spending and especially with restrictions in public sector pay growth. 2.1 Fertility and the Ageing Population Like most European countries, the UK has low fertility and an ageing population. However, the extent of these two trends is less pronounced than in many other countries. After a period of decline in the 1980s and 1990s, birth rates in the UK are on the increase. Following the baby boom in the early 60s, declining birth rates saw the total fertility rate 1 fall from a high of 2.95 children in 1963 to below the replacement rate of 2.1 children, reaching a low of 1.63 in The five years since 2001, however, saw consecutive increases in fertility, with the latest figures showing a rate of 1.86 children in 2006, the highest level since Birth rates have increased at all ages, but particularly to women in their late 30s and early 40s, and there has also been a rise in the proportion of births to women born outside the UK, increasing from 12.8% in 1996 to 21.9% in The period of rising birth rates follows the increasing generosity of child-contingent benefits and maternity provision. Between 1 The total fertility rate measures the average number of children that would be born to a hypothetical woman over her lifetime if she experienced the age-specific birth rates of a particular year across her own childbearing cycle.

10 1999 and 2003, government spending per child grew by 50 percent in real terms, a change that was unprecedented over the previous thirty-year period, while paid maternity leave increased from 14 weeks in 1994 to 39 weeks from Brewer et al. (2007) show that the advent of WFTC alone raised the birth rate of less educated mothers by 10% from 1999, which suggests that policy changes have made a sizable contribution to increasing the UK s fertility rate. Twenty years ago, nearly 80 percent of women with children were in married couples; today the figure is only just over 50 percent. The proportion of lone mothers has risen to nearly 30 percent, reflecting both rising separation and divorce rates, as well as an increase in never-married lone mothers. Alongside this, there has been an increase in the proportion of mothers living in cohabiting couples. In 2007, the majority of babies were born outside marriage for the first time. Rising birth rates mean that the UK population is growing as a result of natural change, as well as through net immigration. However, increasing life expectancy and the ageing of the 1960s baby-boomers will cause the population to age. The proportion aged 65+ is predicted to increase from 15.8 percent in 2000 to 25.3 percent in 2050, although this increase is smaller than that for the EU25 from 15.8 percent to 29.5 percent. Life expectancies at birth, currently 76.9 years for men and 81.3 years for women, are above those for France, Germany, Belgium and Denmark, but below those for Ireland, Italy, Spain and Sweden. 2.2 Labour Market and Unemployment Policy Unemployment had two marked peaks in 1986 and 1993, on both occasions this was around 12% of the economically active. Since 1993 open unemployment fell sharply until 2002, since which time it has been broadly flat at just over 5%. Employment in the UK is high at just under 75% of the working age population (age 16-59/64), a figure exceeded in Europe only in the Scandinavian countries. Even before the fall in unemployment the large numbers reliant on welfare benefits who were not looking for work (i.e., living on sickness and disability benefits and lone parents on Income Support social assistance benefits) became a government priority after the 1997 election. With the decline in open unemployment, this hidden joblessness became a major social issue (see figure 2).

11 The ambition of the incoming government was to raise economic activity and reduce welfare dependency, whilst simultaneously reducing poverty. There are obvious potential conflicts in this strategy. Raising the generosity of benefits reduces the depth of poverty, but is likely to reduce incentives to work. The government s strategy to meet these twin and often contradictory goals consisted of five core elements: Making work pay raising incentives to work Case management by a personal advisor - client engagement and activation within the welfare system undertaken by a personal advisor Limited conditionality welfare payments have attached behavioural requirements, with minimum income secure for vulnerable groups, (e.g., Educational Maintenance Allowance, New Deals for the unemployed, Work Focused Interviews and Work Related Activity Premium/Job Participation Payment under pilot). Addressing key areas of social exclusion that damage people s future life chances (including child poverty, youth long-term unemployment, teenage pregnancy, homelessness etc.) Raising incentives for self-protecting savings among low income groups savings gateway, stakeholder pensions, Children Trust Funds (saving pot schemes, which are also a form of targeted conditional financial support) The first three elements will be discussed in this sub-section, the latter two will be discussed under section 2.6 Poverty and Social Exclusion. Making Work Pay National Minimum Wage Upon its election the current government made a number of key strides to raise takehome pay for low earning jobs. The most prominent measures were the introduction of, and later increases in, the National Minimum Wage and the expansion of the Tax Credit system. The UK had had a system of industry specific minimum wages set by Wages Councils dating back to As industries changed in importance many sectors were combined, but new sectors such as security remained uncovered. After 1979 the value of minimum wages set by Wage Councils declined, and they were

12 finally abolished in In 1998 the Low Pay Commission set a new NMW for the whole country and all sectors, but the initial coverage was low by international comparisons. Indeed, even this level was not raised much above prices until 2001, after which a sharp increase in the minimum wage occurred, but only until The core message of the impact of the NMW is set out in Figure 3. This shows wage growth at each percentile of the wage distribution compared to that at the median (50 th percentile) for two periods: 1992 to 1997, prior to the NMW s introduction; and 1997 to 2003, which covers the introduction in 1998 and substantial increases in relative value in 2002 and Prior to introduction there is pretty much a continuous upward sloping line with twists at the top and bottom reflecting far slower growth in earnings in lowest decile and faster growth in the top decile. By contrast, after introduction the pattern is U-shaped, with the most rapid growth in earnings at the lowest paid part of the distribution, with growth in earnings 20% above that at the median (growth at the median was slightly faster in the second period than the first). The upper half of the distribution has continued in a very similar fashion as before. Note that the faster growth in earnings in the second period is evident all the way up to the 40 th percentile, which is unlikely to be due to the NMW and probably reflects the tighter labour market. However, the impact of the NMW on the lowest paying 8 to 10% of jobs is marked. Figure 3 Growth in Hourly Earnings by Percentile (compared to the median) Source: Annual Survey of Hours and Earnings 35 Increase in Percentile Gross Hourly Earnings Excluding Overtime Minus the Increase in the Median Percentile of the Hourly Earnings Distribution for Adults Aged 22 and Over

13 Making Work Pay Tax Credits Minimum Wages raise the pay of all low paid workers, but take no account of family structure, such as the presence of children or numbers of earners or hours worked in a family. Hence the relationship between low hourly pay and low household income even among working families is far from close. Likewise, the working poor often do not have a worker in the lowest decile or so of hourly earnings. The government made clear its concern about poor work incentives for benefit dependent families with children in a series of publications (The Modernisation of Britain s Tax and Benefit System, numbers 1 through to 5, 1997 to 1999). The response was a major expansion of the country s Tax Credit system (then called Family Credit) under the banner of the Working Family Tax Credit (WFTC) and then later split into two parts Working Tax Credit and Child Tax Credit. Before the 1998 Budget, support for children came from four sources: a universal per- child transfer (Child Benefit) normally paid to mothers, extra payments in means-tested benefits (Income Support) normally paid to the household head in workless families, a refundable tax credit for working families (Family Credit) paid to the mother, and one of two related non-refundable tax credits available to an earner within a couple. 2 The Government has increased the generosity of all four of these starting with the March 1998 budget, and all but Child Benefit have had some form of structural reform. The increases in the generosity of Child Benefit in the 1998 and 1999 budgets together raised the real level of support by 27 percent for the eldest child. The increases in support for children in means-tested benefits was focused on younger children between April 1997 and 2001, such that weekly payments for children up to 4 years of age rose by a week above inflation a 73% real increase whilst support for children aged rose by just 4.25 in real terms. The result is that the level of financial support for children of all ages up to 15 have been equalized older children had previously received more generous support. This reform partly reflects 2 There are approximately 7 million families with children in the UK, and 13 million children.

14 recognition by the Government that poverty rates were higher among families with younger children, and partly it facilitates the move to an integrated child credit. The new Children s Tax Credit was a non-refundable tax credit that replaces two mutually exclusive and equal-valued tax credits: the Married Couple s Allowance (MCA) and the Additional Person s Allowance (APA). 3 The overall impact was that married couples without children lose an extra tax allowance, and families with children, regardless of their marital status, receive around twice as much as before. Whereas the MCA and APA were available to all taxpayers, the Children s Tax Credit was withdrawn at 6.7 percent from people paying higher-rates of income tax (over 33,515 in April 2001). The Working Families Tax Credit (WFTC) was an evolutionary reform to the existing in-work benefit called Family Credit. It was announced in the Labour government s first full budget in spring 1998, and became available to claimants from October The WFTC was available to families where any adult member is working 16 hours a week or more. It consists of a per-family element the same for couples and lone parents and per-child elements. There is a flat zone where the maximum award is paid, and the credit is phased-out beyond an earnings limit of a week at a rate of 55 percent of after-tax income. This equated to 38% of gross income for a person on the basic-rate of tax and National Insurance rates, or a total marginal deduction rate of 69%. WFTC had 4 major differences from its predecessor. First, it was more generous, as both the family and the child elements have risen. For a family with one child, the WFTC was worth a maximum of a week. Each subsequent child raised the maximum credit by a week. Worth noting here is that, although the level of out-of-work support has also risen, these changes by themselves made little difference to work incentives. Second, families can earn more before support is withdrawn. The maximum weekly earnings before withdrawal starts was raised from 80 under the old Family Credit system to Third, the phase-out rate was lowered from 70 to 55 percent of after tax income. These three changes increased 3 The UK has an individual system of income tax. These credits and allowances appear in a person s tax code through which employers assess and then deduct income tax directly from pay checks. Allowances are typically less generous than in the US, so people start paying income tax at lower annual incomes (see Gale (1997) and Brewer (2000) for more comparisons of the US and UK tax systems).

15 support for those in full-time or better paid part-time work (i.e., earning more than 92.90) and extended eligibility to in-work support to a large number of families. Lastly, the WFTC can help with childcare costs though a new Childcare Tax Credit, which paid parents 70 percent of childcare costs up to a (generous) maximum of 100 a week for 1 child or 150 for more than 1 child. The Childcare Tax Credit is restricted to households where all parents are in paid work, but lone parents are the prime beneficiaries to date. The effect of these changes are shown in Figure 4a&b, which show the 2001 and 1997 support packages for a couple with one child. The increased generosity of the WFTC over the previous in-work benefit accounts for 2.7 billion of this 7.2 billion early reform package total, with 2.5 billion going to child payments in means-tested and universal benefits, and 1.8 billion on the Children s Tax Credit (see HMT, 2000). Figure 4 Picture of Welfare Support available for Families with Children under the pre-1997 system (Figure a) and WFTC circa 2001 system (Figure b) Figure a Figure BUDCONa. Pre-WFTC system of support for families with children 160 Level of family support ( /week) Married Couple's Allowance Family Credit Income Support Child benefit Gross earnings ( /week) Note: Assumes entitlement for WFTC reached at 59.20, or 16 hours work at the minimum wage. Values uprated to prices.

16 Figure b) Figure BUDCONb. System of support for families with children from Level of family support ( /week) Children's Tax Credit Working Families' Tax Credit Income support Child benefit Gross earnings ( /week) Note: Assumes entitlement for WFTC reached at 59.20, or 16 hours work at the minimum wage. The impact of the increased generosity of the WFTC and the Children s Tax Credit is muted for those with medium to high rents. Low-earning renters are eligible for rent support known as Housing Benefit (owner-occupiers normally get no housing support). Housing Benefit is phased out as income rises, but at a faster rate than WFTC 65%, and starting at a lower income level. 4 Under FC, the maximum marginal deduction rate could reach 97 percent. 5 The increased generosity of the WFTC compared to FC has meant that fewer households claiming in-work support are eligible for HB, and the reduction in the taper has eliminated the worst of the poverty traps. The effect of the multiple phase-outs, and the effect of the increased generosity in WFTC increased the range of incomes (or rents) over which people are on both WFTC and HB phase-outs has been sharply reduced. This comes at the cost of a 4 All low-income households (not just renters) can apply for rebates for the local taxes in the UK, which are assessed against property values. This rebate Council Tax Benefit is administered alongside Housing Benefit, and phased out at an additional 20 percent. So if families receive both benefits, the combined withdrawal rate on post-tax income is 85 percent. 5 Marginal deduction rate is the term used in the UK for the combined effect of taxes and benefit withdrawals (or phase-outs).

17 large increase in the numbers with phase-out rates at around 70% stemming from the combination of income tax, National Insurance and tax credit withdrawal. In 2003 a second round of reform occurred and the WFTC was split into two parts: a Working Tax Credit paid to the principal earner, and a Child Tax Credit paid to the principal carer. The later now included all the child payments inside Income Support and the tax system (Childrens Tax Credit) and is paid alongside child benefit so that all child payments were now paid to the primary carer through a single system. The new system had a number of other changes, including being assessed over annual income and paid through the year. Last year s income is used as a prediction of this year s annual income and this is then reassessed at the end of the year, and any adjustments (through over- or under-payment) are then made. To avoid too many people repaying credits a disregard is employed, such that if income rises, but remains below this disregard, no repayment is required. This was initially quite low (around 3,000), but has since been raised to over 20,000. The second major change is that the withdrawal has been switched from an after tax net withdrawal rate of 55% to a gross (before tax) rate of 39%. The new structure is laid out in Figure 5 on the same basis as Figure 4 for a family in Figure 5 The Post-2003 structure of Support for Families with Children Figure ICC. The new integrated system of support for families with children 160 Level of family support ( /week) Employment tax credit Income support Integrated child credit Child benefit Gross earnings ( /week) Note: Assumes entitlement for employment tax credit reached at 59.20, or 16 hours work at the minimum wage. Values deflated to prices.

18 Making Work Pay Tax and National Insurance The third element in making work pay entailed a significant reduction in income tax and NI contributions for lower earning individuals. This was done by extending the amount a person could earn before making NI contributions, and introducing a low starting rate of Income Tax at 10p in the pound. These early reforms have now largely been undermined since 2003, as tax- and NI-free allowances were linked to prices, not earnings, and the 10p starting rate of tax was abolished in 2007, whilst reducing the standard rate of tax to 20p from 22p. Case management with Limited conditionality From 1986, under the ReStart programme, the idea of job search conditions for the unemployed was enhanced and monitored with benefit sanctions if breached. This was greatly enhanced in 1996 under the move to Jobs Seeker Allowance (JSA). This renaming of unemployment benefits reflects the extra requirement to look for work actively of face sanctions. Claimants are required to keep a diary of job search activities, interviews etc. and they can be quizzed about this activity every two weeks, although in practice only those about who are thought suspicious are checked in depth. Claimants can not restrict the distance they have to travel to a job or the type of job sort after 13 weeks. This tightening of conditions has undoubtedly contributed to declining unemployment since 1986, but some of the displaced claimants have ended up on other, mostly sickness and disability, benefits. At this time there was limited support for job search activities for the unemployed. What was available included loans for things that would help secure work and provision of short courses in preparing a CV or interview techniques, etc. However, there was nothing for all other claimant groups. The strong conditionality meant a clear line between the unemployed and other claimant groupings (the sick and disabled and lone parents). New Deals for the Unemployed The incoming government added a New Deal structure on top of the required activity explained above. For young people (aged 18-24) after 6 months of claiming JSA, the claimant was placed in an intensive job search regime with more monitoring and help and guidance with job search. This is called the Gateway phase. If after 10 months

19 the claimant is still without work, then they must partake in a designated activity programme. There are four major options: subsidised work, an education or training course that lasts for 9 months to a year, placement with a charitable body or environmental project or a self-employment start up. The decision of which branch to move onto is made in discussion between an advisor and the claimant, however, the education option is restricted to those with lower skills and the subsidised jobs are often in short supply. For older claimants a similar process starts at 18 months, but the activity programme is more flexible in that people can move from one option to another, as the personal advisor deems most appropriate. New Deals for Lone Parents and Sick and Disabled - Personalised Welfare-to-work Support The slightly later New Deals for those on inactive benefits offered a rather different structure. Participation for these groups was voluntary and so a more flexible approach to the support on offer and engagement between the personal advisor was created. This new approach is sometimes described as Personalised Welfare-to- Work Support. If the claimant agrees to pursue participation in the New Deal, an Action Plan is designed delivering a package of support services tailored to the individual s needs brokered by the Case Worker. The Action Plans set out an agreement between the claimant and a case worker on a return to work strategy and time scale. These contain a range of potential support services which are tailored to individual circumstances (short skills courses, confidence building, work trials, condition management, help getting child care, etc.) Evaluation suggests that the New Deal for Lone Parents raises participants moves into work by 22 ppts (Dolton et al. 2005). Work Focused Interviews Upon introduction, participation rates were low in these New Deals for Lone Parents and the Disabled, but they were often successful in helping participants into work. So it was decided to introduce a Work Focused Interview (WFI) whereby Lone Parents (from 2001) had to meet their case worker on a regular basis to discuss the potential for return to work. This will include better off calculations for specimen jobs, discussions on childcare and the Action Plan based support. Attendance for Lone

20 Parents is required, i.e., sanctions can apply if a claimant fails to attend. They will shortly amount to 6 monthly and will be even more intensive when the child turns 14. Evaluations suggest that the WFI regime raised participation in New Deal for Lone Parents by 25% (Knight et al. 2004). There is a danger that such expansion could lead to reduced success in moving claimants into work, as the expansion includes less job ready or less willing participants. Figure 6 suggests that this was not the case with WFIs. The figure plots the numbers participating and numbers entering work after 6 months on the New Deal. They track well with a fairly constant entry rate of about 50%. Note the increase in participation after WFIs were introduced in mid Figure 6 Numbers Participating in the New Deal for Lone Parents and Entering Work within 6 months of starting a New Deal, 1999 to 2004 Source Department for Work and Pensions 1 0 0, , , , , , ,00 0 S ta rters J o bs R a tio 3 0, , , ja n - ju n 9 9 ju l- d ec 99 ja n - ju n 0 0 ju l- de c 0 0 ja n - jun 0 1 ju l- de c 0 1 ja n - ju n 0 2 ju l- d e c 0 2 ja n - ju n 0 3 juld e c 0 3 ja n - ju n 0 4 ju l- d e c 0 4 The New Deal for Disabled People was very similar to that for lone parents, but with extra support for condition management to help people work with their conditions rather than waiting for the condition to improve. It had even lower take up than that for lone parents but again, for those that did the outcomes were good. So following the use of WFIs for lone parents this approach was tried for the sick and disabled in pilot programmes, ones discussed in more detail below.

21 Policy Reform and Employment Figure 2 above charted how the numbers of claimants for the major out-of-work benefits fell from over 6 million in 1995 to 4.7 million in JSA unemployment benefits fell rapidly with the economic upswing from 1996 to 1999, and this decline accounts for 1.2 million of the fall. However, this stopped and reversed a little after The contribution of the New Deal programme to this fall is minor, as the main point is to address long-term unemployment and even then the contribution of the NDs is minor (van Reenan 2000). However, the making work pay strategy has made a larger contribution to reducing joblessness, especially among families with children. The numbers of Lone Parents on Welfare have fallen from 1,025 to 777 thousand (1995 to 2006) and the employment rate of this group has risen from 44% to 57% over this period. This is an extraordinarily rapid rise. Evaluation has consistently shown that the WFTC contributed around 4 percentage points to that rise between 1999 and 2003 (around 70,000) and policy is likely to have had continued effects since the move to the new Tax Credits (see Gregg et al for a summary). Figure 2 above also shows that over this period there was a slight rise in the numbers claiming sick and disability benefits although this peaked in 2004 and has begun to fall. This group, along with lone parents, remain a major policy concern for the government to meet its targets on employment and poverty. Policy development for this group is discussed further below. Current Policy Pilots and Reform Proposals The process of welfare reform in the UK since 1997 has been a rolling one rather than a single big bang. Further, policy makers have regularly sought evidence of some degree of success from reforms in welfare-to-work programmes before national introduction. There has thus been a culture of piloting reforms. The analysis above highlighted some of the remaining important challenges for welfare reform in the UK, the large numbers of disability benefits, the slowing down of employment gains for lone parents, the aging population and (not mentioned in detail) the large numbers of working poor families with two or more children.

22 Overall economic inactivity rates have fallen only marginally to around 21%, but this broadly static picture of inactivity masks an ageing population and rising numbers in full-time education, which push inactivity up. Improvements within key groups, such as lone parents, older workers and the emerging pathways evidence for the sick and disabled, suggest there have been improvements of substantial worth. Figure 7 shows inactivity rates by age (16 to 70) for the UK in 1997 and 2006, excluding fulltime students. The picture shows that there have been marked falls in inactivity amongst those aged 50+, typically of around 5 percentage points, but we see little change among younger groups (around 1ppt declines between 25 and 50). The rise in education participation among the young and the general ageing of the population (in particular, the first post-war baby boom,1947 to 1950, who are now approaching 60), mask this development in aggregate data. Figures 1 and 7 therefore highlight the three (not mutually exclusive) target populations the government wishes to raise employment among: lone parents, the sick and disabled and those over 50. These groups must be brought back into the labour market if the government is going to meet its targets of reducing child poverty (by half the 1998 rate by 2010) and of raising the working age employment rate to 80% (no specified time). Virtually all jobless families with children are now either on lone parent or disability benefits rather than JSA for the unemployed.

23 Figure 7 Inactivity (excluding full-time students) by age, 1997 and 2006 Inactivity by age Percentage Age Current Pilots To address these issues the government has three major current pilot programmes in the field which are producing information for the design of possible policy reforms in 2008 and beyond. The three pilots are: Pathways The Pathways programme seeks to help claimants of sickness and disability benefits (Incapacity Benefit and Income Support with Disability Premium) get back to work. It consists of six Work Focused Interviews in the first 9 months of a person s claim. At these interviews the client is encouraged to join the Choices programme (which can contain mixtures of rehabilitation and New Deal for Disabled People) and is supported by a Back-to-Work Bonus and a Job Preparation Premium. These help make sure work pays and that participants are able to meet any costs associated with the extra activity required. The Choices support package is almost always delivered by an outside contractor, rather than the government agency.

24 Pathways, to date, has applied only to new claimants, but an additional pilot for the short term stock, that is, those who have been in receipt of benefit for between nine months and 3 years, has started. The pilots or Pathfinder areas have been chosen and implemented in three waves. The first covering four relatively small Job Centre Plus districts started in October The second wave started in April 2004 and a more substantive group, which brings coverage to around a third of the country, was implemented from October 2005 through to October The Programme will be rolled out nationally by the end of Pathways appears to be having an impact on numbers leaving incapacity benefits. Figure 8 shows the proportion of initial claimants in a month who are no longer in receipt of benefits 6 months later. Each month reported thus reflects the month of entry on to benefit. The count is of the proportion of the initial claimants who are no longer in receipt 6 months later, so any claimant who leaves but later returns to being a claimant counts as a benefit recipient. It is clear that both phases of Pathways implementation have reduced the numbers of claimants by 7% after 6 months. There are also indications that this decrease has been associated with a reduction in the total number of benefit recipients in the pilot Districts. There is a risk, however, that it is the most job ready who move into work and hence these people would have returned to work anyway, but at a later date (see Adam et al for evidence on Pathways to Work). 6 Current Pilot areas: Began October 2003: Bridgend & Rhondda, Cynon, Taff, Derbyshire, Renfrewshire, Inverclyde, Argyll & Bute; Began April 2004: Lancashire East, Essex, Gateshead & South Tyneside, Somerset PBR04 announcement: Phase 1 Jobcentre Plus Districts (October 2005): Glasgow, Cumbria, Lancashire West; Phase 2 Jobcentre Plus Districts (April 2006): County Durham, City of Sunderland, Greater Manchester Central, Liverpool & Wirral, Swansea Bay West Wales, Lanarkshire & East Dunbartonshire, Barnsley, Doncaster & Rotherham, Tees Valley; Phase 3 Jobcentre Plus Districts (October 2006) Staffordshire, Greater Mersey, Eastern Valleys

25 Figure 8 Percentage of Claimants Leaving Sickness and Disability Benefits after 6 months under the Pathways Programme 50% 45% 40% National Phase 1 Phase 2 Pathways I.B. six month off-flow rate Off-flow rate 35% 30% 25% 20% Apr-01 Jun-01 Aug-01 Oct-01 Dec-01 Feb-02 Apr-02 Jun-02 Aug-02 Oct-02 Month of benefit start The off-flow rates presented are produced from the Working Age Statistical Database (WASD). WASD does not include a proportion of short-term Incapacity Benefit claims, therefore the off-flows presented will be lower than actual rates. However, trends over time will be consistent. Dec-02 Feb-03 Apr-03 Jun-03 Start of phase 1 pilots Aug-03 Oct-03 Start of phase 2 pilots Dec-03 Feb-04 Apr-04 Jun-04 Aug-04 Oct-04 Dec-04 So the crucial question is: how long can these reductions in caseload persist? Figure 9 shows the same information as Figure 8, but 1 year after claims have started, rather than after 6 months. The first monthly cohort of Phase 1 pilots to reach a year from the pilot start are shown by the vertical line and are still showing a 5% reduced caseload on welfare. So, whilst there had been some decay, there was still a marked lowering of caseload after a year. The Phase 2 pilots had not been running for a full year for this data period, but the pre-effect is apparent in the two months prior to start up. Early evidence suggests that the policy is not helping those with mental health problems (e.g., depression and anxiety), nor those with physical conditions.

26 Figure 9 Percentage of Claimants Leaving Sickness and Disability Benefits after 12 months under the Pathways Programme Proportion of customers leaving IB w ithin 12 months 65% 60% 55% Off-flow rate 50% 45% 40% 35% 30% 25% National Phase 1 Phase 2 Start of Phase 1 pilots 20% Month of benefit start The off-flow rates presented are produced from the Working Age Statistical Database (WASD). WASD does not include a proportion of short-term Incapacity Benefit claims, therefore the off-flows presented will be lower than actual rates. However, trends over time will be consistent. The effects of Pathways and the increasing focus on disability within JC+ is for the first time leading to a pronounced decline in the numbers of claimants. Numbers on both disability benefits had previously risen continuously, from around 700,000 in 1979 to a little over 2.77 million in November 2003, but they are now 100,000 below this peak. This represents healthy progress in tackling the problem. New Deal plus for Lone Parents The pilots addressing potential further support mechanisms for lone parents include an in-work credit of 40 on entry into work paid for one year and, in a reduced number of areas, a Work Search Premium designed to generate greater participation in NDLP and other programmes. More recently a full package of potential support including the in-work credits and an activational payment (entitled Work Related Activity Payment WRAP) has started, labelled the ND+LP pilot programme. Early evaluation of this new package suggest that it has raised lone parent employment in targeted areas by 1.2 ppts, but it had a more marked impact in raising employment among those with over 12 months of claiming benefits previously (where the change was 7ppts after 2 years of the scheme).

27 Employment Retention and Advancement Pilot The Employment Retention and Advancement pilot (ERA) is aimed at understanding better what policy mix can support retention of re-entrants in the labour market after leaving welfare benefits. This focuses on time-limited financial support and follow up contact between the case manager, the worker (former claimant) and their employer. In addition, this pilot programme is trying to engage with non-claimant/non-working partners of those in-work, but at risk of in-work poverty. The first evaluation evidence is now available and a few pointers are clear. Engaging non-working partners proved very difficult. The ambition to improve job retention has seen reductions in repeat benefit claims for both New Deal groups (NDLP and ND 25+) of 4-5 percentage points. Earnings advancement has a 29% increase in earnings for NDLP group (mainly by raising working hours rather than hourly wages with a 7pp increase in likelihood of working full-time), but much smaller effects on earnings for WTC/ND25+ groups. Skill development has also proved somewhat difficult, but an increase in likelihood of combining training/education with employment of 14 percentage points for lone parents on WTC, but much smaller for other groups. Compared to US evidence for such programmes, this does appear encouraging. 7 Looking Forward New Disability Benefit A Welfare Reform Bill currently in parliament suggests a new test for disability benefit claimants. The two different disability benefits (one contributory and one means-tested social assistance) are to be merged into one. Those making claims will be split into two groups: the most disabled will receive the full benefit without conditions, whilst those with less severe conditions (based on functional capabilities) will only receive the full benefit if they engage in the action plan process described earlier under Pathways. If they fail to meet the agreed activity requirements, they will lose part of their benefit (perhaps 30, which would take them done to the same rates as JSA). 7 DWP briefing, sourced from Dorsett et al (2007) Implementation and first-year impacts of the UK ERA demonstration, DWP research report 412.

28 Further Lone Parent Reforms A proposed further reform for lone parent benefits is to transfer lone parents on to JSA unemployment benefits when the youngest child reaches 12 and in due course to age 7 (rather than 16 at present). They will then be treated as a special category of the JSA unemployed, with extra support services to help the transition back to work. Therefore, this will be rather like the New Deals for the Unemployed, but the details are not yet clear. These developments would move the UK much further toward a mixed approach of support services and sanctions, whereas to date conditionality has been used rather lightly. 2.3 Old-age Pensions The population s ageing is putting increased strain on the UK pension system. Unlike many other OECD countries, there is not the prospect of rapidly growing state spending on pensions, but the pension system is not without its problems. The first tier of the UK pension system is the Basic State Pension, a flat-rate, fairly minimal pension paid to men from age 65 and women from age 60, rising to 65 from Receipt of the full Basic State Pension depends on contributions and there are big gaps in coverage for those with caring responsibilities, with more than twothirds of women retiring today failing to qualify for a full pension. Since the early 1980s, the Basic State Pension has been up-rated annually in line with inflation (rather than wage growth) and its value has fallen relative to earnings. On top of the Basic State Pension sits a second tier of compulsory pension provision. The default is membership of the state scheme (formerly the state earnings-related pension scheme SERPS, now the state second pension S2P), but individuals can choose to opt out into a private pension. When it was introduced, SERPS was intended to pay a pension worth one-quarter of an individual's best twenty years' earnings (up to a specified upper earnings limit), but its value has gradually been reduced and in the medium term, S2P will become a flat-rate top-up to the basic state pension. Even at its most generous for individuals retiring at the turn of the century, however, the combined pension from BSP and SERPS yielded a

29 replacement rate of around 40 percent for someone on median earnings. This is relatively low by European standards. Rather than increasing the value of contributory pensions, the Labour government chose to target financial support at the poorest pensioners by raising the value of means-tested benefits through the introduction in 1999 of the Minimum Income Guarantee, changed to Pension Credit in These reforms were relatively successful in reducing pensioner poverty and raising replacement rates at the bottom of the income distribution, although take-up rates of Pension Credit are still quite low estimated at around 60 percent of caseload. Since their introduction, the value of means-tested benefits for pensioners has been annually up-rated in line with earnings and, as their reach has crept up the income distribution, there has been growing concern about the potential disincentive effects on saving that such extensive means-testing might have. This was identified as a key issue by the independent Pensions Commission reporting on the state of the UK pension system in One of its central recommendations was to restore the earnings link in the Basic State Pension in order to reduce dependency on means-testing, and this is due to be implemented in There will also be more generous credits for those with caring responsibilities. Compared to many other European countries, the UK has a relatively high level of private pension provision. Between 1978 and 1988 individuals could opt out of the state second tier into a defined benefit (final salary) scheme. From 1988 onwards individuals could additionally opt out into a defined contribution (money purchase) pension scheme. The option to opt-out of the state system helped to preserve occupational DB pensions and encouraged take-up of individual DC schemes. Nearly half of people now entering retirement receive more money from one of these private (occupational or individual) schemes than they do from the state. However, the relatively low and falling generosity of the state schemes means that the overall system is very reliant on private schemes to deliver reasonable replacement rates to the middle and top of the income distribution. Moreover, there is concern about growing gaps in coverage of private schemes that will leave people with relatively low levels of income in retirement.

30 Cost pressures from increasing longevity, as well as from a growing burden of regulation, have caused many DB occupational schemes to close to both new and (albeit less frequently) existing members. Take-up of individual DC schemes has failed to fill the gap and, if anything, has also been in decline. The government introduced stakeholder pensions from 2001 to provide a regulated, but privatelyprovided individual DC scheme suitable for low- to middle-earners, but their introduction has had almost no effect on overall take-up rates. The government s policy towards private provision was initially one of informed choice, underpinned by a belief that if given the right opportunities, people will plan ahead sensibly. In other words, the government would provide information and make suitable products available, but leave individuals to make savings decisions themselves. However, there is a growing body of evidence from new behavioural economics that letting people make their own decisions does not necessarily yield optimal outcomes; that default options, for example, can have a huge effect on whether or not people belong to a pension, how much they save and where they choose to invest. The conclusion drawn from this literature is that the government can and indeed should encourage saving by framing individuals pension choices. The result has been a move to introduce a new National Pension Savings Scheme which will automatically enrol all eligible employees into a low-cost individual pension account, albeit with the option to opt out. Individuals in the scheme will be required to contribute a minimum of 4% earnings, with employers contributing a further 3%, and the government, 1%. The introduction of automatic enrolment through this scheme, recommended by the independent Pensions Commission, and taken up by the Government in its recent White Paper, is intended to boost take-up of private pension saving. The third main proposal from the Pensions Commission which has been adopted by the government concerns gradually raising the state pension age for men and women from 65 in 2020 to 68 by This follows a period of over twenty years from the 1970s to the mid 1990s during which retirement ages have fallen, in practice to well below the current state pension ages. Together with rising life expectancy, this has meant that people now spend a longer proportion of their lives retired (see figure).

31 Figure 10 Years spent in education, work and retirement, by date of birth cohort Note to figure: Retirement age for 1950 and 1980 cohorts is assumed to be unchanged from 1935 cohort. Age when leaving school and retirement age based on Family Expenditure Survey. Data on life expectancy from ONS Population Trends (2004). In raising the state pension age, the government wants to signal the need for a behavioural change, for if we are living longer, we may need to work for longer. The UK already has a lower inactivity rate among older workers (aged 55-64) than much of continental Europe and the rate has fallen by 5 percentage points since 1997 to around 30%. In practice, what the government does to the state pension age is likely to have only a limited direct effect on employment among this group, since retirement decisions are likely to be more affected by incentives in private pensions than what is happening to the state pension. However, much of the evidence points to people working into older ages in the future. Until recently, many DB schemes encouraged early retirement, but now, faced with growing deficits, they are reducing provision for early retirement and raising normal retirement ages. So, retirement ages among people with occupational pensions will undoubtedly rise in the future. For people with individual schemes, the age at which they retire will depend on the size of funds that they have been able to build up and